MigrationNarrate: A Dataset for Detection of Migration Narratives in YouTube Videos

Researchers have introduced MigrationNarrate, the first multimodal dataset specifically designed for detecting migration narratives in YouTube videos within the UK context. This development addresses a significant gap in the study of migration narratives, particularly in video-centric public communication platforms. The dataset facilitates the analysis of how migration is framed in online discourse through multimodal signals, which is critical for understanding societal perceptions and communications.

Why it matters

This research introduces a critical tool for analyzing how migration is portrayed and understood in modern digital communication. The ability to detect and analyze migration narratives on video platforms allows for deeper insights into public opinion, societal trends, and the framing of complex social issues, which can inform strategic communication and policy development.

Key insights

• Narrative detection is crucial for understanding public discourse and social communication framing.
• Migration narratives are significantly understudied, largely due to a lack of dedicated annotated datasets.
• Public communication has increasingly shifted to video-centric platforms where narratives are conveyed multimodally and consumed at scale.
• Narrative analysis in video content remains largely unexplored.
• MigrationNarrate is introduced as the first multimodal dataset for detecting migration narratives in UK YouTube videos, comprising 1,115 entries.
• The dataset aims to bridge the gap in understanding migration narratives within the context of video-based online platforms.

Source

arXiv — Computers and Society — https://arxiv.org/abs/2608.20984